gao-16-188, federal real property: gsa could decrease ... · for all nine of the leases gao studied...

FEDERAL REAL PROPERTY

GSA Could Decrease Leasing Costs by Encouraging Competition and Reducing Unneeded Fees

Report to Congressional Requesters

January 2016

GAO-16-188

United States Government Accountability Office

United States Government Accountability Office

Highlights of GAO-16-188, a report to congressional requesters

January 2016

FEDERAL REAL PROPERTY GSA Could Decrease Leasing Costs by Encouraging Competition and Reducing Unneeded Fees

Why GAO Did This Study More than half of GSA’s 377 million square feet of space were leased from the private sector as of 2014. While GSA strives to match or beat private sector leasing rates, it is important to identify any opportunities to increase efficiency and reduce costs.

GAO was asked to review GSA’s leasing costs. This report examines (1) the extent to which GSA’s leases achieve market rates and how overall federal leasing costs could be reduced and (2) how GSA's cost estimates for selected leases compared with the actual costs of leasing paid by federal tenants. GAO determined the extent to which the rates of a sample of 714 GSA leases compared to market rates; analyzed selected leases for office space across all 11 GSA regions in more detail; and interviewed officials from all GSA regions and 2 GSA tenant agencies, as well as private sector real estate representatives. GAO contracted with a real estate consultant for the market rate analysis.

What GAO Recommends GSA should (1) enhance competition by encouraging tenant agencies to modify their geographic and building requirements; (2) explore seeking authority to use Federal Building Fund balances to reduce interest fees; and (3) give tenants the option to reduce fees by choosing non-cancelable occupancy agreements. GSA agreed to increase competition and determine if it can use Fund balances to pay tenant improvement costs but disagreed with allowing tenant agencies to choose non-cancelable occupancy agreements. GAO believes GSA should provide this option as a potential cost-saving measure.

What GAO Found GAO found that the General Services Administration’s (GSA) lease rates exceeded comparable market rates for many of 714 leases reviewed. Specifically, a review of these leases from 2008 through 2014 determined that about half exceeded their local market’s average private sector rate for similar space by 10 percent or more. According to officials from all 11 GSA regions and private sector stakeholders, GSA is unable to more consistently achieve lower rates because competition among private lessors for these leases is limited; this limited competition is due to factors including tenant agencies’ requested geographic areas and specialized building requirements, as well as the length of GSA’s leasing process. For example, an agency’s initial requested geographic area may be so restricted that it does not include any buildings that meet all tenant requirements, resulting in increased costs and time as GSA explores alternatives.

In addition, overall federal leasing costs increase when tenants finance needed improvements to newly leased space—called tenant improvements—over time. GSA tenants routinely amortize these costs over the term of their leases and pay interest rates of up to 9 percent to the building’s owner. Because GSA’s tenants lack sufficient upfront capital, they chose to amortize their tenant improvements for all nine of the leases GAO studied that included those costs. In total, these 9 leases will incur $15 million in interest fees to be paid to private owners—nearly 40 percent of the total paid for these tenant improvement costs. GSA manages a fund—the Federal Buildings Fund, which pays rent and other real property costs—with sufficient unobligated balances to loan tenants enough funds to cover tenant improvement costs and avoid paying private sector interest, but GSA does not have budget authority to fund such costs. GSA also requires most of its tenants to sign cancelable occupancy agreements, which permit tenants to vacate leased space under certain circumstances in exchange for a higher fee paid to account for the risk of GSA’s possibly having to find a new tenant for the space. However, the importance of routinely including this built-in flexibility for short term leases is questionable, as it is not often exercised. Allowing tenants the option of choosing non-cancelable agreements would reduce tenant fees.

The actual leasing costs paid by tenant agencies exceeded GSA’s estimates for 7 of the 11 leases finalized from 2000 to 2014 that GAO reviewed in more detail. Seven of those leases were “standard” leases (costing less than $2.85 million in annual rent, as of fiscal year 2014) and four were “high value” (costing more than $2.85 million). For 4 of the 7 standard leases tenants’ actual leasing costs exceeded GSA’s estimates by more than 10 percent. Inaccurate estimates complicate tenant agencies’ planning, but tenant agencies often have to accept increases in GSA’s cost estimates because some lack authority to independently lease space. GSA officials said that the lack of competition for GSA leases and changes to tenant agencies’ space needs during the leasing process contribute to cost growth. Conversely, GAO found GSA’s initial cost estimates for 4 high-value leases to be more accurate than those for standard leases. High value leases, which represent only 4 percent of leases but more than 40 percent of GSA’s leasing costs, are subject to congressional authorization, which may help control cost growth.

View GAO-16-188. For more information, contact Dave Wise at (202) 512-2834 or [email protected].

http://www.gao.gov/products/GAO-16-188


mailto:[email protected]

Page i GAO-16-188 GSA Lease Estimates

Letter 1

Background 3 GSA Could Better Achieve Market Rates and Reduce Tenant

Costs for Its Leases by Fostering Competition and Reducing Fees 8

GSA’s Cost Estimates for Selected Leases Were More Accurate for High-Value Leases, Which are Subject to Congressional Approval 21

Conclusions 26 Recommendations to GSA Administrator 27 Agency Comments and our Evaluation 27

Appendix I Objectives, Scope, and Methodology 30

Appendix II Comments from the General Services Administration 40

Appendix III Comments from the Social Security Administration 42

Appendix IV GAO Contact & Staff Acknowledgments 43

Table

Table 1: Eleven GSA Leases Selected for Analysis of GSA Leasing Cost Estimates 37

Figures

Figure 1: Square Footage of Real Property Held or Leased by the General Services Administration (GSA), 2001 through 2014 4

Figure 2: Description of Typical Steps for the General Services Administration’s (GSA) Acquisition of Real Property Leases 5

Contents

Page ii GAO-16-188 GSA Lease Estimates

Figure 3: Variation among 11 Regions between General Services Administration (GSA) Rents and Market Rents for Comparable Properties, By GSA Region, 2008 through 2014 10

Figure 4: Firm Terms of 4,285 GSA Leases Sampled for GAO Analysis, Executed between 2008 and 2014 15

Figure 5: Total Tenant Improvement Costs if Amortized over Term or Paid at the Outset of a Lease for Three Selected General Services Administration (GSA) Leases Executed between 2000 and 2014 18

Figure 6: Actual Leasing Rates for Seven Selected Standard Leases as Percentage of Initial General Services Administration’s (GSA) Initial Estimates, 2000 to 2014 22

Figure 7: Actual Leasing Rates for 11 Selected Standard and High-Value Leases, as a Percentage of General Services Administration’s (GSA) Initial Estimates, 2000 to 2014 23

Page iii GAO-16-188 GSA Lease Estimates

Abbreviations DOJ Department of Justice FBF Federal Buildings Fund GSA General Services Administration IREM Institute of Real Estate Management MSA Metropolitan Statistical Areas NPV net present value OMB Office of Management and Budget SSA Social Security Administration

This is a work of the U.S. government and is not subject to copyright protection in the United States. The published product may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately.

GAO-16-188 GSA Lease Estimates

441 G St. N.W. Washington, DC 20548

January 13, 2016

The Honorable Ron Johnson Chairman The Honorable Thomas R. Carper Ranking Member Committee on Homeland Security and Governmental Affairs United States Senate

The Honorable John Barrasso United States Senate

The federal government’s management of real property remains on GAO’s high-risk list due to complex and long-standing issues, including an overreliance on leasing of privately-owned space in situations wherein ownership for stable agency needs would be more cost efficient in the long run.1 Although some agencies have independent authority related to the leasing of real property, the others lease space through the General Services Administration (GSA)— the federal government’s primary civilian real property management agent. As of fiscal year 2014, the most recent year for which GSA published portfolio information, GSA had a total of 377 million rentable square feet in its inventory—slightly more than half of which were leased from the private sector.2 One of GSA’s key strategic goals is to procure leased space from the private sector at or below industry market rates. To this end, GSA has recently embarked on two rounds of leasing reform; however, there may be additional opportunities to improve federal leasing practices and save money. Further, leases are likely to be a stable or growing part of an agency’s portfolio due in part to capital limitations. Thus, it is important to understand the GSA’s leasing process and identify any opportunities that could increase its efficiency and result in federal cost savings.

You asked us to assess the financial implications of GSA’s leasing program. This report examines (1) the extent to which GSA achieves

1GAO, High-Risk Series: An Update, GAO-15-290 (Washington, D.C.: Feb. 11, 2015). 2GSA’s government-owned properties constitute the remaining 183 million rentable square feet in the Public Buildings Service portfolio. GSA Public Buildings Service, FY2014 State of the Portfolio, (Washington, D.C.: 2014).

Letter



market leasing rates for its leases and how overall federal leasing costs could be reduced, and (2) how GSA’s cost estimates for selected leases compared with the actual costs of leasing paid by federal tenants.

To address these objectives, we took a number of steps. To assess the extent to which GSA achieves market leasing rates for its private sector leases, we contracted with a real estate consulting firm chosen through a competitive process to compare a sample of 4,285 GSA leases executed between 2008 and 2014 with private sector leases of similar location, size, and quality in the same major markets during the same period. From the full GSA data set, a sample of 714 GSA leases across all of the agency’s regions was selected from markets with sufficient information. The GSA data were abstracted and compared with data from published brokerage reports in specific markets and submarkets to determine the extent to which GSA received a comparable private-sector market rent for its leases. To assess how GSA’s cost estimates compare with the actual costs of leasing paid by federal tenants, we selected and analyzed documentation for 11 GSA leases—one from each of GSA’s regions. We selected these leases based on criteria including property type, most recent lease action, and rental rate per square foot using GSA data. Our non-generalizable sample of 11 GSA leases—comprised of 7 standard leases and 4 high-value leases that became effective between 2000 and 2014—included properties designated as office space across all GSA regions and among 6 federal tenant agencies. Further, we reviewed documentation, analyzed the leasing process, estimated the actual costs over the active term of the 11 leases selected and interviewed key GSA staff for all selected leases.3 We also reviewed GSA leasing policies and guidance and interviewed officials from GSA headquarters about the lease procurement process. Finally, we interviewed officials from all 11 GSA regional offices, as well as the two tenant agencies with the most leases in our sample—the Department of Justice (DOJ) and Social Security Administration (SSA)—regarding their experiences with the GSA leasing process. For more information on our scope and methodology, see appendix I.

We conducted this performance audit from August 2014 to January 2016 in accordance with generally accepted government auditing standards.

3Two agencies accounted for 6 of the 11 leases reviewed; four were GSA leases on behalf of the Department of Justice and 2 were on behalf of the Social Security Administration.


Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives.

The federal government’s civilian real property holdings include hundreds of thousands of buildings and permanent structures across the country that cost billions of dollars annually to rent, operate, and maintain. Within this portfolio of government owned and leased assets, GSA plays the role of broker and property manager to many civilian agencies of the U.S. government. The Administrator of GSA is authorized by law to enter into lease agreements, not to exceed 20 years, on behalf of federal agencies. The administrator delegates leasing authority to GSA regional commissioners, who further delegate authority by issuing leasing warrants to lease contracting officers.4 GSA manages its inventory via 11 regional offices and its central office, located in Washington, D.C. While GSA’s Office of Portfolio Management is responsible for establishing the strategies and policies for GSA’s real property portfolio, its regional offices are generally responsible for conducting day-to-day real property management activities, including leasing, in their region.

Federal management regulations specify that when seeking to acquire space for an agency, GSA is to first seek space in government-owned buildings and vacant space already under lease to the government. If suitable government-controlled space is unavailable, GSA is to acquire space in an efficient and cost-effective manner. As shown in figure 1, the square footage of property leased by GSA has steadily increased in recent years while the amount of federally owned space held by GSA has remained steady.

4According to GSA, lease contracting officers perform the warranted duties that obligate the federal government, including executing and administering lease contracts and safeguarding the interests of the United States in its contractual relationships. In this role, lease contracting officers are responsible for obtaining all necessary legal, technical, and financial advice within GSA necessary for the sufficiency of lease contracts and for compliance on the part of lessors.

Background


Figure 1: Square Footage of Real Property Held or Leased by the General Services Administration (GSA), 2001 through 2014

The process for acquiring leased space, as outlined in GSA’s Public Building Service Leasing Desk Guide, begins when GSA receives a request for space from a federal agency.5 Using this guide—which provides guidance on implementing federal property regulations—GSA officials then work with an agency to fulfill the specific requirements for the space, including the square footage and any geographic limitations.6 According to GSA guidance, developing and finalizing these details

5The Public Buildings Service Leasing Desk Guide contains authorities, policies, technical and procedural guides, and administrative limitations governing the acquisition by lease of federal real property. According to GSA, the provisions of the desk guide apply to all PBS personnel engaged in the acquisition and administration of lease contracts, as well as agencies leasing space under delegated authority from GSA. 6During this phase, the tenant agency determines, among other things, the necessary square footage, circulation requirements, parking requirements, and the number of personnel to be housed.


should take anywhere from 2 to 8 months, depending on the complexity of a tenant agency’s space needs. After this initial stage, the Leasing Desk Guide estimates that approximately 18 to 24 months are needed to procure new leased space. During this time, GSA takes a number of steps to complete a lease acquisition (see fig. 2). During the lease acquisition process (‘leasing process’), GSA compiles and shares iterative estimates of the leasing costs with tenant agencies pursuing space. Prior to the advertisement step, GSA and each tenant agency involved sign a draft occupancy agreement detailing the estimated costs associated with a lease. At the conclusion of the process—when the actual costs of leasing a specific space are known, following the “build-out and acceptance” step—GSA and the agencies execute a final occupancy agreement associated with a specific lease, which allows agencies to budget for future payments.

Figure 2: Description of Typical Steps for the General Services Administration’s (GSA) Acquisition of Real Property Leases

GSA is required to take additional action for prospectus leases—in 2014, those new leases with a net annual rent7 above $2.85 million. For these leases—also known as ‘high-value’ leases—GSA must submit a prospectus, or proposal, to the House and Senate authorizing committees

7GSA’s PBS Leasing Desk Guide defines net annual rent as total rent minus operating expenses paid directly to the lessor.


for their review and approval.8 Given this additional requirement, GSA’s Leasing Desk Guide suggests the lease acquisition process for high-value leases begin 3 to 5 years prior to lease expiration. The prospectus should include the purpose and location of the lease, as well as basic information about the space to be leased including the location, an estimate of the maximum cost to the government of the space, and a statement of rent currently being paid by the government for federal agencies to be housed in the space. This information assists Congress in overseeing GSA’s management of its real property portfolio. Typically, these prospectuses are drafted in the GSA regional offices and reviewed and approved by GSA’s Office of Portfolio Management. The prospectuses are then reviewed and approved by the Office of Management and Budget prior to being provided to congressional authorizing committees—the Senate Committee on Environment and Public Works and the House Committee on Transportation and Infrastructure. In 2013, we reported that there were 218 active GSA high-value leases, which accounted for about one-third of GSA’s net annual rent costs; GSA’s overall lease inventory included more than 8,300 leases as of August 2015, 4 percent of which had current annual rents above the $2.85 million threshold.

Once GSA executes a lease on behalf of a tenant agency and an occupancy agreement with the agency, that agency is required to pay rent to GSA for the space they occupy. Rent payments are deposited into the Federal Buildings Fund (FBF), which is a fund established by the Public Buildings Act Amendments of 1972.9 Congress provides annual

840 U.S.C. § 3307. According to GSA, leasing specialists make an initial determination about whether a prospectus is required as soon as a tenant agency provides its space requirements. As a matter of policy, GSA officials said that an advertisement for a high-value lease cannot be published until the Office of Management and Budget (OMB) has cleared the prospectus and GSA has submitted the signed prospectus to the House and Senate authorizing committees for their review and approval. Further, GSA's policy is to not enter into high-value leases until the authorizing committees have adopted a resolution approving the project. 9The Federal Buildings Fund is administered by GSA. The FBF is funded primarily by income from rental charges assessed to tenant agencies occupying federally-owned and GSA-held or leased space that approximate commercial rates for comparable space and services. Congress exercises control over the FBF through the appropriations process that sets annual limits on how much of the fund can be obligated for various activities. In addition, it periodically provides supplemental appropriations for the Fund. As of 2012, funds from the FBF supported nearly 10,000 assets—including about 1,500 federally-owned buildings held by GSA and 8,100 leased by GSA—which provided an inventory of more than 370 million square feet of workspace for 1.1 million federal employees.


limits on the amount GSA may obligate to provide a range of real property services.10 As of February 2015, the FBF had an unobligated balance of $3.6 billion.

Included in federal agencies’ monthly rent is a monthly fee to GSA for its services related to leased space; as of 2015, tenants paid 5 or 7 percent of their lease value in a fee to GSA based on the level of flexibility the agency had in canceling the agreement.11 Depending on the extent to which leased space must be altered for an agency to fulfill its mission, the costs of the improvements necessary (“tenant improvements”) are paid by the tenant to the lessor through GSA. Tenant improvement costs include changes to walls, electrical outlets, telephone lines, and secure rooms that need to be made by the lessor between the time that GSA executes the lease and the point when the tenant agency takes occupancy. GSA officials said that it is standard practice for tenant agencies to amortize these costs over the lease term and noted that this approach is similarly utilized in the private sector. GSA assigns each tenant to one of 6 tiers, which equate to the standard improvements that tend to be required to prepare a space to support the mission and activities of a particular agency.

In late 2009, GSA began to reform its leasing by simplifying the lease acquisition process, among other changes. For example, for leases below $150,000 in annual rent, GSA introduced a ‘simplified lease’ model to allow a more efficient way to process documents customized for lower-

10In March 2014, we found that limitations on obligations constrain funding for real property, especially for repairs, alterations, and new construction. At that time, officials from one agency said that GSA had been unable to obtain FBF funding through the annual appropriations process to complete renovations in their federally owned and GSA-held or leased property. Similarly, GSA officials noted at that time that when Congress provides less obligational authority than requested, repairs, alterations, and new construction projects for GSA-held or leased buildings are most affected because available funds are first used to pay leasing, operations and maintenance, and debt costs. (GAO-14-239) 11GSA officials stated that this fee is 7 percent when the ability to cancel the occupancy agreement prior to lease expiration—with a 120 day notice—is included in the occupancy agreement and 5 percent when it is not. Space is deemed to be non-cancellable when there is a low probability that the Public Buildings Service would be able to find a backfill tenant due to specific qualities of the space. GSA guidance states that the agency reviews each space assignment and uses certain specified criteria to designate space as cancelable or non-cancelable.



value leases. GSA included this and other reforms in its Leasing Desk Guide of policies and procedures, which was introduced in April 2011.

While GSA’s fiscal year 2015 budget request stated that one of the agency’s strategic objectives is to procure leased space on behalf of federal tenants at or below market rates, our analysis of recent GSA office leases across all regions performed for this report found that about half of the rates negotiated in recent years exceeded market rates at the time the leases were executed.12 Although GSA analyzes its lease rates against market rates over the full term of the lease, we chose to limit our assessment to the point in time the leases were executed, as this is the moment at which actual market rates were known. In doing so, the review of 714 new GSA leases finalized between 2008 and 2014 found that about half exceeded their local market’s average rate for similar space by

12GSA has reported that, in fiscal year 2013, it executed leases at a weighted average of 10.1 percent below market rate. As stated above, GSA compares its lease rates against market rates over the full term of the lease; we limited our assessment to the point when the leases were executed.

GSA Could Better Achieve Market Rates and Reduce Tenant Costs for Its Leases by Fostering Competition and Reducing Fees

About Half of Recent New GSA Lease Rates Have Exceeded Comparable Market Rates


10 percent or more.13 This phenomenon varied across the 11 GSA regions, with some regions performing better than others. In general, three GSA regions were in line with market values for rents, five GSA regions executed leases that were on average at or below market, and three GSA regions had rates that exceeded local market rates on average.14 Specifically, figure 3 shows how the rates negotiated by GSA regional offices compared to relevant market rates in the years the leases were executed. GSA officials across all regions stated that they utilize a number of tools to establish a range of market rental rates for each lease, including a report specifically tailored for each transaction with market information, analysis, and insight regarding the relevant location.

13This finding is based on the 714 leases abstracted during broader analyses of new and replacing GSA leases and published brokerage reports in the specific markets and submarkets. This analysis included all GSA regions, focused toward Metropolitan Statistical Areas (according to the Office of Management and Budget) within each, where the greatest volume of GSA and market data was available on rents and expenses. This exercise involved multiple leases per year in each market for each year of the analysis period. Leases negotiated prior to the period of 2008 to 2014 were not included in the comparison. It is notable that 3 agencies which accounted for approximately one-third (240) of the full sample had a larger percentage of leases above the market range—defined as being 10 percent above or below the average market rate—than the entire sample. These agencies had a median rate of 15 percent above the market range, while the median of the full sample was 5 percent above the range. When these agencies are removed from the set, the remaining 465 leases have a median within the market range. For more information, see appendix I for methodology. 14The Great Lakes, Heartland, and Greater Southwest regions were generally balanced in performance; the New England, Northeast and Caribbean, Southeast, Rocky Mountain, and National Capital regions were generally at or below market, and the Mid-Atlantic, Pacific Rim, and Northwest/Arctic regions were generally above market. GSA officials told us that spaces in their Pacific Rim and Northwest Arctic regions are subject to seismic requirements and that the agency is only able to lease within buildings that meet such standards.


Figure 3: Variation among 11 Regions between General Services Administration (GSA) Rents and Market Rents for Comparable Properties, By GSA Region, 2008 through 2014

Note: The gray area represents the 10 percent variance “zone” in which GSA leasing rates are deemed equivalent to market rate. Not every GSA region is represented in each of the years because some properties leased on behalf of federal agencies at rates with an extreme variance above or below market rate were removed from the analysis.

GSA could more consistently achieve market rates or better if there were more competition for its leases. The Federal Management Regulation requires that federal agencies acquire leased space at rates consistent with prevailing market rates through full and open competition.15 As we reported in 1995, this is designed to ensure that all responsible sources are allowed to compete and serves as the government’s primary price control mechanism.16 At a June 2015 hearing, a top GSA manager stated that GSA’s ongoing lease reform effort includes plans to reduce costs by

1541 C.F.R. §102-73.55 and 41 C.F.R. §102-73.100. 16GAO, Federal Office Space: More Businesslike Leasing Approach Could Reduce Costs and Improve Performance, GAO/GGD-95-48 (Washington, D.C.: Feb. 27, 1995).

Lack of Competition for GSA Leases Is Driven by Multiple Causes

http://www.gao.gov/products/GAO/GGD-95-48


increasing competition.17 However, according to our interviews with officials across all GSA regions as well as private sector stakeholders, competition among private lessors is currently limited by the following factors:

Restricted geographic area: GSA regional officials said that tenant requirements for the location of a leased space reduced the number of buildings that qualify, thereby limiting competition for GSA leases. In some cases, an agency’s requested geographic area may be so restricted that it does not include a single building that meets all the tenant’s requirements.18 For example, for one lease we reviewed, a tenant agency was forced to twice widen the geographic areas it initially requested in order to find the space ultimately leased. GSA officials from this region said that these starts and stops necessitated by the narrow geographic area requested by the tenant resulted in repeating steps of the leasing process, increasing related costs and the overall schedule. Although officials from GSA’s central office told us that GSA has not typically questioned the appropriateness of an agency’s location limitations, GSA regional officials said that broadening the area deemed suitable for potential properties—while still ensuring that the agency can pursue its mission—is one of the best ways to increase competition for a lease. In September 2015, an internal GSA memo to all regions detailed a renewed policy through which GSA is taking the initiative to suggest geographic areas to tenant agencies. This memo states that GSA will consult with regional officials on several factors in order to designate geographic areas and, when agencies submit future space requests to GSA, this policy will help to enhance competition.

Specialized building requirements: GSA regional officials said that competition for GSA leases is further limited by federal agencies’ specific building requirements. These requirements—including things such as

17Statement by Commissioner of Public Buildings Service, U.S. General Services Administration at U.S. Senate Homeland Security and Governmental Affairs Committee Hearing, “Federal Real Property Reform: How Cutting Red Tape and Better Management Could Achieve Billions in Savings,” (Washington, D.C.: June 16, 2015). 18GSA’s Leasing Desk Guide states that the geographic area requested by an agency, known as the ‘delineated area,’ is defined by the specific interconnecting boundaries within which space will be obtained to satisfy an agency space requirement. This area is identified by the tenant agency during the requirements development process, based on location policy requirements found in the Federal Management Regulation and the agency’s mission needs.


parking spaces, ceiling heights, or security setbacks—affect the number of qualifying properties left in a position to bid for the GSA lease. GSA regional officials said some property owners do not want to make the investment in their building to meet a tenant agency’s security requirements when they can choose to lease their space to a non-federal tenant. In the previously mentioned case wherein a tenant agency was forced to twice widen the geographic areas it initially requested to find space, the tenant agency’s space request included elevators and a large square amount of footage on the first floor, which played a key role. GSA could not meet these requirements in the original geographic area requested, and alternatives including strip malls or warehouses were not viable for the agency. As a result, GSA had to look more broadly across the market to identify a qualifying space that met all of the agency’s requirements.

Presence of unique clauses: Private sector leasing representatives and regional GSA officials also said that clauses in GSA leases not found in private industry leases make them less competitive. For example, private sector officials we spoke with told us that the substitution clause—which gives GSA the unilateral right to substitute any other tenant for the original intended tenant—deters some landlords that do not want to risk the possibility that the replacement agencies may not be compatible with the existing tenants. At a workshop hosted by GSA in June 2015, GSA officials said that they were considering changes to the substitution clause, which they said causes risk with lenders and financiers. However, in October 2015, GSA officials told us that they would not consider changes to the clause without proof that it increases costs or harms building owners.19 Regardless, the flexibility which such a clause is designed to offer GSA and its federal tenants is not regularly exercised, according to GSA data. Further, officials from one GSA regional office explained that, while compliance with state laws also exists in a private sector environment, they also have to comply with policies specific to federal buildings. Other clauses mentioned by GSA regional officials as being unique to GSA leases include federal energy efficiency and

19GSA officials told us that it is unlikely that the substitution clause would be adjusted or removed from the standard GSA lease template; they stated that they studied the possibility, but said that changing this clause would be counterproductive to GSA’s larger space management role. They added that they do not feel that there is a major risk associated with the substitution clause because the private lessor would always have a revenue stream through GSA, and a tenant would be in the space.


conservation clauses; private sector representatives also cited GSA’s fire and casualty clause, which immediately terminates a GSA lease if a building in which space is located is totally destroyed by fire or other casualty. The presence of these additional requirements could also make renting to private sector tenants more desirable for some property owners.

Lengthy leasing process: The lengthy GSA leasing process—when compared to the private sector—also has the potential to reduce competition for GSA leases. From the date of GSA’s initial cost estimate to the point when an agency took occupancy, the 11 GSA leases we reviewed took, on average, almost 4 years to complete, with some taking as long as 6 to 8 years.20 GSA took more than 2 years to complete the leasing process for 10 of the 11 leases we reviewed, and building owners receive no rent during this protracted process, causing some owners drop out of the competition and others to likely choose not to bid knowing the long time frames involved. Officials from GSA headquarters told us that it is important to note that the amount of time to execute a lease is only one of several variables for assessing GSA performance and shorter lease execution timeframes are not always better; it could, for example, reflect a rushed, poorly planned process.21 However, the extent of GSA’s administrative obligations and paperwork can cause delays. For example, a regional GSA official said that the time required for federal tenants to approve architectural drawings often exceeds that for private sector tenants and, while some more sophisticated lessors will include the costs of these delays in their price, this can deter some potential lessors from even submitting a bid.

Officials from GSA’s central office said that although GSA’s goal is to meet or beat private sector leasing rates, federal leases are different than private sector leases and some of the differences can make it difficult to

20Of the 11 GSA leases for which we conducted an in-depth documentations review, 4 were high-cost leases and 7 were standard GSA leases. The length of the leasing process for the high-value leases, which require congressional authorization, ranged from 2.4 to 8.6 years and was 5.4 years on average. The length of the process for the 7 standard leases ranged from 1.1 to 5.7 years and took 3.0 years on average. 21GSA stated that variation in leasing contract officers’ approaches could play a role and explained that taking more time to complete the process can be advantageous, adding that proactive planning can result in changes being less expensive and the interaction with the tenant agency being more pleasant.


compete with private sector leasing rates. Specifically, these officials stated that:

• GSA must procure leases based on specific award factors governed by federal law, a process that can discourage competition for these leases or cause lessors to price their risk accordingly; decisions related to private sector leases are more flexible and could be based on individual preferences alone, thus making them less cumbersome and more desirable to some property owners.

• For a federal lease, a property owner must sign a GSA-drafted lease document and abide by its requirements; conversely, the contractual instrument used in private sector lease is provided by the owner and crafted according to its terms.

• While a private sector tenant generally must leave a property when a lease expires, federal agencies often continue to occupy leased space after the expiration of a lease term, often in holdover status without the contractual right to occupy the space if the government and lessor are working out details or disagree.

GSA officials told us that as a result of these differences, federal leases are often more complex than private sector leases for some property owners, and the pricing of that complexity and business risk for private owners can subsequently translate into greater expense for federal agencies.

GSA officials said that increasing the term of GSA leases is a key part of GSA’s efforts to reduce leasing costs, but our analysis found that longer terms do not necessarily lead to lower costs in the first year of leases. Based on our analysis of the agency’s data, GSA typically negotiates relatively short-term leases—that is, those with guaranteed terms fewer than 5 years in length. While GSA considers 80 percent of the 4,258 leases we reviewed to be 10-year leases or longer, many of these leases have a 5-year guaranteed (“firm”) term followed by an optional (“soft”) term. The private sector views leases structured in this way as 5-year leases because that is the only part that is guaranteed. Figure 4 shows that 70 percent of the new GSA leases we analyzed, finalized from 2008

GSA Has Opportunities to Reduce Federal Tenants’ Leasing Costs GSA’s Focus on Longer Lease Term May Not Result in Cost Savings but Could Offer Other Benefits


through 2014, had firm terms of 5 years or less. Conventional wisdom—according to both GSA officials and private sector real estate professionals—holds that shorter term leases are typically more costly.

Figure 4: Firm Terms of 4,285 GSA Leases Sampled for GAO Analysis, Executed between 2008 and 2014

However, based on our sample, it is not clear that GSA leases with shorter firm terms actually do cost more than those with longer firm terms. In the last year, GSA has been encouraging agencies to obtain longer leases with a 10-year firm term where appropriate and, in June 2015, a GSA manager testified that GSA plans to extend lease terms to 10 years or longer in order to reduce costs.22 However, our analysis of new GSA leases executed on behalf of federal agencies from 2008 through 2014 found no direct financial benefit based on the length of the firm term of a

22Statement by Commissioner of Public Buildings Service, U.S. General Services Administration at U.S. Senate Homeland Security and Governmental Affairs Committee Hearing, “Federal Real Property Reform: How Cutting Red Tape and Better Management Could Achieve Billions in Savings,” (Washington, D.C., June 16, 2015).


lease in the base year that the leases were signed. 23 This lack of cost savings is attributable, in part, to greater tenant improvement costs that are more prevalent in longer-term leases.

While it may not reduce costs directly, increasing the number of leases with longer firm terms, as GSA plans to do, could offer other benefits to GSA. While GSA officials in the regions said that leases with firm terms of 5 years or less provide flexibility to tenant agencies that may not need the space for long periods, few agencies take advantage of the flexibility of the 5-year lease. The average length of time that federal agencies remained in space leased through GSA was more than 23 years—possibly through multiple occupancy agreements—for the GSA leases that expired between 2001 and 2014. In addition, given GSA’s lengthy leasing process, short-term leases cause challenges for GSA and tenant agencies. For the 11 leases we reviewed in depth, the average time needed to complete a GSA lease was nearly 4 years (an average of 3 years for standard leases and more than 5 years for high-value leases), ranging from more than one year to more than 8 years. Further, this process may soon take longer: GSA officials said one aim of GSA’s ongoing lease reform is to begin the process even earlier—at least 36 months before the expiration date for standard leases.24 This would mean beginning the process shortly after the start of an agency’s initial occupancy of a space for a 5-year lease.

As stated earlier, new leases often involve costs related to the customization of the space known as tenant improvements, which are usually amortized over the term of the lease.25 Tenant agencies can fund these costs in two ways: (1) pay for the improvements at the outset, prior to moving into the space, when negotiations between GSA and a property

23Although we did not find that term affects the lease rate in the first year, GSA’s rates may improve relative to market rates over the term of an individual lease based on any included rate escalations and the performance of the local real estate markets. 24GSA officials stated that the aim for high-value (‘prospectus-level’) leases is to begin the process 60 months prior to lease expiration. 25As stated earlier, tenant improvement costs include changes to walls, electrical outlets, telephone lines, and secure rooms that need to be built by the lessor between the time that GSA executes the lease and the point when the tenant agency takes occupancy of the space.

Financing Agencies’ Needed Improvements to Leased Space over Time Increases Federal Leasing Costs


owner permit26 or (2) amortize the costs of the improvements over time during the lease financed by the building owner. GSA regional officials said that nearly all tenants choose to amortize their basic tenant improvements over the firm term of the lease, and the analysis of GSA leases performed for this report supports this assertion. Nearly 60 percent of leases in our full data set of 4,285 leases involved tenant improvement costs—all of which opted to amortize at least some of these costs over a period during the lease. Both GSA and commercial real estate firms tend to amortize the costs needed to prepare a leased space for tenant occupancy over the firm term of the lease and ask landlords to assume the risk—GSA doing so on behalf of its tenant agencies—of customizing a space according to specific requirements. Because private owners that lease to the federal government assume this responsibility and obtain the resources required to construct, operate, and maintain real property over the course of its lifecycle, federal agencies then pay private sector interest rates as they pay for their improvements over the firm term of their GSA lease.

The overall cost of leasing office space increases considerably when agencies opt to amortize their tenant improvement costs over time instead of paying them at the outset. When agencies amortize their tenant improvements during their lease, they pay substantial sums to private lessors in the form of interest based on the rates GSA negotiates with private lessors on agencies’ behalf. In this approach, tenant agencies pay not only the sum of the principal and interest, but also additional GSA fees—either 5 percent if they are in a non-cancellable occupancy with GSA or 7 percent if they are in a cancellable agreement—typically over the firm term of the lease. Nine of the 11 leases we reviewed had tenant improvement costs and more than one-third of the costs related to these improvements were directed toward interest, as all 9 cases amortized these costs. These 9 leases incurred an average of $1.7 million in interest costs related to tenant improvements. In total, these 9 leases incurred a total of $39 million in tenant improvement costs, of which nearly 40 percent ($15 million) was due to interest paid to private lessors. For example, in one lease we reviewed, the tenant agency chose to amortize its $2.1 million of tenant improvement costs over the life of a 15-year lease at a 9 percent interest rate, which will ultimately cost $4.0 million

26One tenant agency stated that GSA does not always permit them the option to make a lump sum payment for their tenant improvements; they said this is dependent upon the stage of GSA’s negotiations with the property owner.


after including both the $1.7 million to be paid in interest charges and GSA’s 5 percent fee on those charges. The agency could have saved 45 percent, more than $1.8 million, over the term of its GSA lease if these costs had been paid at the outset. Additional examples from our analysis are illustrated in figure 5.

Figure 5: Total Tenant Improvement Costs If Amortized over Term or Paid at the Outset of a Lease for Three Selected General Services Administration (GSA) Leases Executed between 2000 and 2014

Although agencies typically lack the resources to fund improvements at the outset of a lease according to GSA officials, there may be opportunities to reduce overall federal leasing costs by identifying funds to reduce the amount of interest paid to private lessors. The Federal Management Regulation states that the basic real estate acquisition


policy is to acquire real estate in an efficient and cost-effective manner.27 We have previously reported that lack of capital to finance real property investments, including tenant improvements, has been a long-standing challenge for GSA and other federal agencies. However, identifying sources of capital to fund tenant improvement costs at the outset would reduce federal agencies’ leasing costs. One possible option to reduce the costs paid by tenant agencies could be to provide budget authority for GSA to finance the capital needed for tenant improvements to be paid at the outset of a new lease and have the tenant pay it back over the term, without the interest charges a tenant agency currently pays. For example, it is possible that GSA could use available balances from the Federal Buildings Fund (FBF) to fund tenant improvement costs, with sufficient controls in place, at the outset of a lease. The FBF is administered by GSA and was established in 1972 as the primary source of funds for operating and capital costs associated with federal space. GSA collects rent from tenant agencies, deposits it into the FBF, and uses that money—as authorized by Congress—to fund real property acquisition, operation, maintenance, and disposal. The FBF has contained unobligated balances for several years and, as of February 2015, the fund had an unobligated balance of $3.6 billion. However, GSA does not currently have the budget authority to use the unobligated balances in the FBF to fund tenant improvements. GSA officials said that the concept of funding agencies’ tenant improvements using unobligated FBF balances has potential, but also said that GSA has not formally considered this approach. They said that applying unobligated balances in this way has the potential to save substantial amounts money on interest charges that are currently passed onto federal tenants, but that the risks and opportunities would need to be fully studied.

GSA also requires most tenants to sign cancelable occupancy agreements, which can also increase federal leasing costs for agencies and may not be needed. GSA regularly requires tenant agencies to sign cancelable occupancy agreements that allow them to vacate the leased property under certain circumstances. Non-cancelable tenant agreements require the tenant agency to pay rent on the leased property for the entire firm-term of the lease. GSA charges more in administrative fees (7 percent of total rent instead of 5 percent) for cancelable tenant agreements to account for the higher risk of having to replace a tenant

2741 C.F.R. §102-73.10.

GSA Requires Agencies to Pay More for Flexibility They May Not Need


before the end of a lease. However, according to GSA’s Pricing Desk Guide, GSA does not allow agencies to decide whether the cancelable agreement warrants the higher fee; rather, GSA reviews each leased space and determines whether to designate its agreement as cancelable or non-cancelable. Officials from GSA headquarters told us that their regional officials determine whether or not an agency’s agreement should be non-cancellable and, further, these regional officials do so based on whether they think they will be able to find a replacement tenant, not on the tenant’s likelihood of canceling. For example, GSA officials hypothesized that it may be difficult to find a replacement tenant for a Transportation Security Administration leased space located beyond the security line at an airport, thus the tenant’s agreement with GSA would likely be non-cancelable.

However, the importance of routinely including the right to cancel in short-term leases is questionable and we believe that tenant agencies are in the best position to decide how to best meet their consolidation objectives. As mentioned earlier, the Federal Management Regulation states that the basic real estate acquisition policy is to acquire real estate in an efficient and cost-effective manner.28 GSA officials said that most agencies agree to pay the additional 2 percent in management fees in exchange for the flexibility that it gives them to have a cancelable agreement. However, officials from two GSA tenant agencies we interviewed said that this built-in flexibility can be useful, but they also said that they rarely exercise the right to cancel their agreements with GSA. Officials at GSA headquarters also said that tenants rarely exercise their right to cancel their occupancy agreements; as one official at a GSA tenant agency explained, this is often because their agency usually considers its space requirements when they are already nearing lease expiration. Moreover, according to GSA, 83 percent of GSA leases were connected to at least one cancelable agreement as of July 2015 while about 70 percent of the more than 4,200 GSA leases considered in our broader analysis have a firm term of 5 years or fewer. Based on GSA’s policy of starting the leasing process from 18 months (standard) to 5 years (high-value) before a lease expires, an agency seeking to cancel its agreement would need to spend a substantial portion of a 5 year period working to arrange their move to a different space, reducing the likelihood that they would cancel early.

2841 C.F.R. §102-73.10.


The actual costs of selected standard leases we reviewed generally exceeded GSA’s initial cost estimates. The reasons for these overruns, discussed later in this section, included lack of competition for GSA leases and changes in agencies’ space needs during the leasing process. As shown in the figure, six of the seven standard leases we reviewed exceeded their initial rent per square foot cost estimates and four of these exceeded their estimates by more than 10 percent; overruns ranged from 6 percent to 90 percent greater than the estimates GSA provided to its tenant agencies.

GSA’s Cost Estimates for Selected Leases Were More Accurate for High-Value Leases, Which are Subject to Congressional Approval Federal Agencies’ Actual Costs Generally Exceeded GSA’s Estimates for Selected Standard Leases


Figure 6: Actual Leasing Rates for Seven Selected Standard Leases as Percentage of Initial General Services Administration’s (GSA) Initial Estimates, 2000 to 2014

aSee sidebar.

GSA’s cost estimates for standard leases have limited oversight mechanisms. Regional GSA officials indicated that the primary requirement for increasing the estimated leasing costs is updating the occupancy agreement with the tenant agency. Officials from the two GSA tenant agencies we interviewed said that GSA’s initial estimates often vary from the final cost; officials from one of these agencies told us that GSA’s early cost estimates are usually significantly lower than true market leasing rates. This can complicate agencies’ ability to effectively plan their budgets and identify other sources of funding after making decisions based on GSA’s initial leasing cost estimates. However, the officials also said that they are unable to pursue other options, as they do not have independent leasing authority to procure similar leases without the assistance of GSA.

Explanation of standard lease #7

Several factors played a role in the 90 percent increase from GSA’s rate estimate to the actual leasing rate paid by the tenant agency for standard lease #7 we reviewed. In 1996, the tenant agency requested a new lease with a 10-year firm term, which GSA estimated would cost $25 per square foot. However, the agency’s stated space requirements—including parking—ultimately necessitated a build-to-suit space, which was not awarded until June 2001 at a rate of $47.42 per square feet. GSA officials stated that the process was stopped at this point and should have been issued a new project number. Because the lease was awarded 5 years after the initial cost estimate, market conditions were subject to fluctuation. Further, GSA’s estimate was made prior to the prevalence of third-party data sources, which help to facilitate more accurate leasing cost estimates. Source: GAO. | GAO-16-188


Among the 11 GSA leases we reviewed in-depth, GSA’s initial cost estimates were more accurate for high-value leases that require a prospectus subject to congressional authorization than the cost estimates were for standard leases. In fact, as figure 7 shows, agencies’ actual leasing costs per square foot were within 10 percent of GSA’s initial estimates for 3 of the 4 high-value leases that we reviewed.

Figure 7: Actual Leasing Rates for 11 Selected Standard and High-Value Leases, as a Percentage of General Services Administration’s (GSA) Initial Estimates, 2000 to 2014

GSA regional officials said that the prospectus process required for high-value leases requires congressional authorization, which increases their accountability for developing accurate initial estimates because if actual costs were to substantially exceed the estimated costs approved in the prospectus, GSA would have to obtain additional budget authority from Congress. The actual costs of three of the four high-value leases we reviewed were slightly below GSA’s estimates. While the final rental rate for one of the high-value leases exceeded GSA’s per square foot estimates by about 20 percent, GSA kept the total cost of the project within authorized costs by reducing the overall amount of space it

GSA’s Cost Estimates Are Better Aligned with Actual Costs for Selected High-Value Leases, Which Require Congressional Authorization


leased.29 Nonetheless, requests for additional budget authority from Congress for high-value leases remain rare. GSA officials said that the agency does not track the number of amendments made to authorized high-value leases or how many have been resubmitted to Congress since 2000, only noting that it has rarely, if ever, happened.

Both GSA officials and these results suggest requiring congressional authorization for high-value leases may help control cost growth for these leases. The congressional approval requirement may be more appropriate for high-value leases because of the larger dollar amounts involved. Although just 4 percent of GSA’s overall lease inventory had current annual rents above the $2.85 million threshold as of August 2015, they account for 44 percent of GSA’s total net annual leasing costs.

According to both GSA regional officials and representatives of GSA tenant agencies, factors similar to those that played a role in causing higher rates for GSA leases can also cause costs to exceed GSA’s initial estimates:

• Officials from GSA regional offices stated that a lack of competition for GSA leases can directly contribute to variations between initial cost estimates and actual leasing costs. As stated earlier, limited numbers of qualifying properties and the reluctance of some building owners to bid for GSA leases can limit competition. When fewer offers are made by owners of appropriate space, GSA regional officials said that initial cost estimates are less likely to align with actual leasing costs. For example, officials from one GSA region told us that the time when offers come in from the market after the solicitation is when rental rates fluctuate the most—and is when the ultimate rental rate could fall on either end of the range set forth in the initial estimates.

• The lengthy GSA leasing process also leaves GSA and federal agencies vulnerable to increased leasing costs. As stated earlier, the 11 GSA leases we reviewed took almost 4 years from the point the lease acquisition process started to completion on average, with some

29Specifically, the prospectus documentation originally included 3 federal agencies in a facility as large as 314,000 square feet but resulted in a 233,000 square feet facility for the FBI alone for nearly the same cost. GSA’s Leasing Desk Guide, published after this lease was executed, prohibits a tradeoff of square footage for rental rate and major changes to the housing plan—both of which occurred on this lease.

Similar Factors to Those That Drive Up Costs of Leases—Lack of Competition and GSA’s Lengthy Leasing Process—Can Cause Actual Leasing Costs to Exceed Initial Cost Estimates


standard leases taking from more than 3 years to nearly 6 years for GSA to fulfill all agency requirements to ready the space for occupancy. The longer this process lasts, the more local lease markets and inventory can change, reducing the reliability of initial estimates. Officials from one GSA tenant agency said that the leasing process can be made longer in some cases because GSA is currently limited to seeking existing space on behalf of agencies. These officials suggested that also allowing the consideration of new construction could increase competition for GSA leases and potentially reduce the need to re-advertise lease requests on agencies’ behalf.

• Changes to agencies’ space needs during the GSA leasing process can also negatively affect the accuracy of initial cost estimates and hinder GSA’s negotiating power. According to GSA’s leasing guide, each tenant agency is obligated to plan and budget for its upcoming space needs 18 to 36 months before the space is needed, further lengthening the period of time over which agencies’ needs can change.30 For example, officials from a GSA tenant agency said that their staffing levels and rental requirements can fluctuate during the typical 24 to 36 months of GSA’s lease acquisition process and that such changes can increase space requested and increase costs. GSA also uses federal agencies’ initial leasing specifications to estimate tenant improvement costs, but the actual prices are negotiated after the lease is awarded. As a result, GSA regional officials said that unanticipated, tenant-driven changes can add time and increased costs to a project even after a lease is awarded. Officials from that tenant agency said that GSA’s leasing cost estimates prove to be more accurate for projects that do not experience any significant delays or changes in scope.

Officials from one GSA tenant agency told us that when leasing costs exceed GSA’s initial estimates, it creates budgeting challenges—not only because it can force them to identify additional sources of funding when costs are greater than initially anticipated, but also because there is not a reliable pattern in terms of variance between GSA’s estimates and their actual leasing costs. However, as stated earlier, tenants contribute to some of the causes that result in actual leasing costs exceeding GSA’s initial estimates.

30GSA’s Leasing Desk Guide states that leases involving new construction require additional time and agencies planning to pursue a high-value lease procurement may need to begin 60 months in advance of their planned occupancy.


Leasing has comprised a growing share of GSA’s portfolio over the last 15 years and will likely continue to remain a significant part of the federal property management system. While it is preferable to own property if an agency has stable rather than variable needs requiring flexibility, it is also important to reduce leasing costs. GSA’s goal is to lease space at or below local market rates, but we found that GSA paid rates 10 percent or more above market rates in the first year about half of the time from 2008 through 2014. GSA has taken steps to reform leasing—and those efforts continue. As part of its ongoing leasing reform efforts, GSA hopes to reduce its lease rates by increasing the length of the terms of its leases, but increasing the competition for GSA leases may yield more benefits. GSA regional officials who negotiate the agency’s leases said that increasing competition for GSA leases is key for achieving market rates, but circumstances often combine to drive down competition. Factors such as a tenant agency’s need for space in restricted geographic areas and specialized building requirements can limit the number of properties that qualify. Reducing barriers to competition, where possible, could increase the number of property owners bidding on GSA leases and, in turn, contribute to GSA more consistently obtaining lease rates at or below market rates.

GSA could also reduce the costs federal tenants pay for their leases. GSA tenants often pay high interest charges—up to 9 percent, among leases we reviewed—to finance their finishes, known as tenant improvements. Identifying sources of capital to allow tenants to fund tenant improvements at the outset of their leases could reduce these specific costs related to leasing by a third—saving millions of dollars for some leases. The Federal Buildings Fund (FBF) has unobligated balances that, with sufficient controls, could help fund tenant improvements. However, further study of the risks and opportunities are needed before GSA seeks additional budget authority from Congress to use FBF balances in this way. GSA could also reduce leasing costs for its federal tenants by allowing them to the option of committing to stay in their space for the full term of their agreement in exchange for a lower fee, particularly for short leases. Currently, GSA requires most tenants to sign cancelable occupancy agreements. However, it takes tenants years to plan and budget for a move, and more than two-thirds of the GSA leases we reviewed have firm terms of 5 years or less, reducing the likelihood of their canceling occupancy agreements early.

Conclusions


As part of its lease reform efforts and to increase possible cost savings, we recommend that the GSA Administrator take the following steps:

1. Fully explore strategies to enhance competition for GSA leases by encouraging tenant agencies to broaden their allowable geographic areas and to limit their specialized building requirements to those justifiably unique to the federal government.

2. Seek to reduce leasing costs for federal agencies by:

• Exploring, with relevant stakeholders, the possibility of loaning unobligated Federal Buildings Fund balances to agencies to cover tenant improvement costs that would otherwise have to be financed for new leases. If GSA finds that, with sufficient controls in place, tenant improvements can be safely funded this way, it should participate in the development of a legislative proposal to request that Congress make the necessary budget authority available.

• Allowing tenant agencies the option of choosing non-cancelable occupancy agreements with lower administrative costs, particularly for leases with firm terms of 5 years or less.

We provided a draft of this report to GSA for review and comment. We also provided a draft of this report to two GSA tenant agencies that we spoke with during our review: the Social Security Administration (SSA) and the Department of Justice (DOJ). GSA provided written comments that are reprinted in appendix II. SSA provided a letter, reprinted in appendix III, stating that it had no comments on our report. DOJ provided technical clarifications, which we incorporated where appropriate.

GSA agreed with the recommendation to fully explore strategies to enhance competition for leases by encouraging tenant agencies to broaden their allowable geographic areas and limit their specialized building requirements to those justifiably unique to the federal government. GSA stated that it has been working aggressively to maximize competition in its leasing program and is increasing national oversight and support for lease planning, project management, and procurement activities. With regards to our recommendation to seek to reduce tenants’ interest costs by exploring the possibility of loaning unobligated FBF balances to agencies to cover tenant improvement costs that would otherwise have to be financed for new leases, GSA agreed to evaluate its existing authorities to determine if the FBF could be used to fund tenant improvements.

Recommendations to GSA Administrator

Agency Comments and our Evaluation


GSA did not agree with the recommendation to allow tenant agencies the option of reducing administrative fees by choosing non-cancelable occupancy agreements, particularly for leases with firm terms of 5 years or less. GSA stated in its letter that it is responsible for assigning and reassigning space to support agency space requirements, balancing risk and flexibility to manage the leased portfolio. GSA’s letter also noted that having the flexibility to return underutilized space is an important tool in meeting the administration’s objective to consolidate space and reduce the federal real property footprint. We agree that agencies should consolidate space as appropriate and contribute to meeting this objective. However, as our review and analysis show, this flexibility costs tenant agencies an additional two percent of the value of the lease, although tenant agencies rarely use it. Even if a tenant agency exercises its right to return space to GSA before the end of a lease, it may be particularly difficult to realize savings for leases with firm terms of 5 years or less; GSA’s 3 year process to plan and execute leases severely limits the extent to which tenants could cancel leases of 5 years or less and any savings would likely be offset by the higher GSA fees these agreements require. Further, in such situations, tenant agencies are given no say in a decision that requires them to pay additional fees. Ultimately, we believe tenant agencies are in the best position to decide how to meet their consolidation objectives. Thus, if agency officials determine that their agency may benefit from the flexibility to cancel all or part of an agreement before the end of the lease, they should be able to choose to pay for that flexibility, rather than being required to do so.

We are sending copies to the appropriate congressional committee and the Acting Administrator of GSA. In addition, the report will be available at no charge on GAO’s website at http://www.gao.gov.

If you or your staff have any questions about this report, please contact me at (202) 512-2834 or [email protected]. Contact points for our Offices


of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made major contributions to this report are listed in appendix IV.

Sincerely yours,

David Wise Director, Physical Infrastructure Issues

Appendix I: Objectives, Scope, and Methodology

Page 30 GAO-16-188 GSA Lease Estimates

Our objectives were to assess: (1) the extent to which the General Services Administration (GSA) achieves market leasing rates for its leases and how overall federal leasing costs could be reduced, and (2) how GSA’s cost estimates for selected leases compared with the actual costs of leasing paid by federal tenants.

To determine the extent to which GSA achieves market leasing rates for its leases, GAO contracted with a professional services real estate firm, selected through a competitive process based on previous experience and cost, to analyze data on GSA and private sector real property leases. The scope of data considered for this objective included information on both private sector and GSA leases commencing between calendar years 2008 through 2014, which were compared and analyzed to determine any differential in lease payments attributable to specific differences in the leases, based on their financial performance. GSA provided GAO with a data set that included all of its active leases from its multiple data sources, which was aggregated using GSA’s unique lease identifiers. The data provided by GSA were reviewed by the contractor for inconsistencies and discussed during interviews with knowledgeable agency officials to assess the appropriateness of their use. GAO also reviewed the contractor’s report, including the steps taken to assess the reliability of the GSA data used and the types of analysis conducted, and found them to be sufficiently reliable. Incomplete and inconsistent data, as well as outliers, were excluded, though accuracy of the individual lease data points provided by GSA was not independently verified.1 The analysis conducted was consistent with how GSA currently handles leases, uninfluenced by agency practices and policies no longer in effect. For example, revisions to processes from the National Broker Contract, Advanced Acquisition leases, and industry outreach sessions have affected GSA practices, thus impacting how leases were negotiated and executed.

Per the data provided by GSA, there were 8,499 GSA leases in effect as of December 31, 2014. Of this population, 5,160 leases commenced after

1The analysis relied upon GSA to provide information on the GSA leases. This lease data were pulled by GSA and provided in 33 separate Excel workbooks. From all files provided, the following contained applicable numeric and descriptive information that was assembled into a single database for all leases. Information was combined using the lease identification number as a primary key, and database queries were developed for further data selection for each analysis.




January 1, 2008, and were initially considered as part of the analysis period. We considered only GSA operating leases—not prospectus level (“high value”)—as well as how the data were dispersed among regions, how many leases are full service versus not full service, how many leases were added to the region per year, and to identify outliers; the result was 4,285 remaining leases matching the following criteria for analysis: classified as 100 percent office space, full service rents, and executed within the dates of analysis (2008 to 2014). These 4,285 leases comprised the universe of leases upon which all analysis was performed. These leases were then broken down into subsets for specific analysis. The pool of leases analyzed for each of the clauses and actions varies based on the compliance of those leases to the criteria for that pool. There was symmetry in the dispersion, again indicating a valid sampling of leases. Misclassified properties were identified within this data set, and outliers were not analyzed or included in the subsequent data sets. The rental rates per square foot analyzed were all-inclusive.

The collection of private sector lease data for markets nationwide for each year of the analytical period included the collection and consideration of:

• Quarterly and annual published market reports for 151 markets and submarkets nationwide, collected from 58 national and regional brokerage companies and used to summarize market rent based on actual leases and listings in each market and submarket.

• Fifteen private sector office leases for national companies, often publically traded and viewed as credit-worthy tenants by brokers and landlords who were parties to the lease.

• Office building operating expenses with detailed line item breakdown for 90 commercial business district and suburban submarkets from the Institute of Real Estate Management (IREM), for urban and suburban buildings of 40,000 to 99,999 square feet and 100,000 to 250,000 square feet.

• Typical lease structures, including base term and options with corresponding tenant improvement allowances and commission structures compiled from broker interviews.

This private sector lease information provided the foundational knowledge of the quantified and qualified performance of private sector office markets in each market over the analysis period. To assess the appropriateness of the private sector data, the contractor reviewed brokerage reports from several sources for each market and conducted interviews with 35 private sector real estate brokers nationwide using a



semi-structured interview protocol, which covered topics such as typical years of a lease, market practices, and responsibility for various expense and service fees. In addition, GAO reviewed the source of the broker market reports and clarified analytical steps with the contractor, including comparability of tenants between the private sector leases and GSA leases. As a result of these steps, the private sector lease data were found to be sufficiently reliable. Grouping of the private sector lease data occurred in several analyses:

• City and submarket analysis to identify a single market rental rate for Class A and for Class B office rents for each year of the analysis for comparison against GSA base year rates.

• Matching of private sector markets within the Top 50 Metropolitan Statistical Areas (MSAs), as defined by the Office of Management and Budget, and markets covered by IREM to identify the portion of rent typically for rent, operating expenses and tenant improvement costs.

• Identification of IREM operating expenses for the same year as the GSA lease base year for comparison between GSA negotiated operating expenses and market operating expenses.

• Identification if GSA leases were equivalent to market, above or below market in each region and each year.

For this analysis, the contracted firm compared the base year contract rent for a GSA lease to the market rent cited from brokerage reports as of the date of lease commencement.2 The number of leases incorporated into each of the data points for GSA leases was a weighted average of the actual rental rates for each lease signed in that year. There were four parts of the data GSA provided that, when added together, resulted in the rental rate paid by GSA: shell rent, general tenant improvements, custom tenant improvements, and operating expenses. The analysis for comparative performance of GSA leases and private sector leases involved:

• abstracting GSA leases based on GSA data, disaggregated to differentiate between shell rental rate, general tenant improvements, custom tenant improvements, and operating expenses;

2The GSA rental rate was the full service gross rent in the base year of the lease.



• considering the net present value of the GSA leases, using the total cost of the lease and payments each year, discounted back at the Office of Management and Budget (OMB) discount rate of 2.80 percent (nominal 10 year rate);

• considering nominal total cost of lease for both GSA and private sector leases; and

• plotting GSA lease and private sector lease performance in each market (such as by size, location, year executed) to identify any variations based on lease terms.

From the full data set, a sample of 714 GSA leases was selected from the MSAs from which we had an abundance of market and operating information, across all GSA regions, to be compared with data from published brokerage reports in specific markets and submarkets to determine the extent to which GSA received a comparable private-sector market rent for its leases. These leases were selected based on the locations within the counties identified within each MSA.3 The match to MSA was to capture the significant published market data available to benchmark. The specific submarket and building class of each lease were matched from the brokerage report, and the difference in rent was calculated between GSA contract rent and the brokerage report market rent. The market rent for each MSA was researched at the city and the submarket level and then further refined for Class A and Class B properties.

Matched pairs of GSA leases were also analyzed for cases in which all factors of leases were highly comparable, except for the length of the term. The characteristics by which leases were matched include the commencement year of the lease, the submarket, building class, and square footage. This analysis involved the isolation of specific lease terms, such as length of term or period of amortization of tenant improvement costs, to determine differentials from the net effective rent, then capturing any variation in the isolated terms. By isolating the single factor and term, differences attributable to this factor would be apparent as all other factors were the same. There were 96 leases matched in

3Twelve major metropolitan areas/cities were selected that represent all 11 GSA Regions. The GSA leases analyzed within these metropolitan areas were ‘new’ and ‘new replacing’ leases for each year of the analytical period. These leases were the basis of the analysis for which a comparison against the private sector rental rates for the same period and same jurisdictions was made.



groups of two and three for locations nationwide. In the matching analysis, termination rights for both 5- and 10-year leases were in place and rental rate was broken out by shell rent, general tenant improvement costs, and custom tenant improvement costs. The first portion of this analysis was based on the base year contract rent against base year contract rent for each pair. As the leases were in the same submarket and of the same size, the premise was that the two properties were competitive and comparable to each other and that the only differentiator was the term of the lease. The second portion addressed the net present value over the full term of the lease, assuming that termination rights were not exercised and that the leases were intact for the complete term. This analysis enabled the reduction in rent after the firm term to be recognized for those leases with amortized tenant improvements.

Further analysis was performed where the firm 10-year term was isolated and compared against a 10-year lease with a 5-year termination right (i.e., a 5-year firm term with a subsequent 5-year soft term). The data was sorted by lease year, then square footage, region and shell rent. The leases were viewed to analyze the difference in net present value (NPV) when the 10-year firm and 10-year with soft term leases were compared, and also when the two leases begin with approximately the same dollars in shell rent or have equivalent shell rent, tenant improvement costs, and operating expenses. A breakdown was created for the leases by region for the number of firm five/soft five versus firm 10-year term for the 10-year leases. The analysis reviewed the entire data set for termination rights and extracted leases with 3-, 5-, and 10-year terms to compare total costs and the discounted net present value of total costs. Further analyses were performed where the firm 10-year term was isolated and compared against a 10-year lease with a 5-year termination right. The leases were viewed to analyze the difference in NPV when 10-year firm leases and 10-year lease involving a soft term were compared and when two leases begin with approximately the same dollars in shell rent or have equivalent shell rent, tenant improvement costs, and operating expenses.

To determine how GSA’s cost estimates compare with the actual costs of leasing paid by federal tenants, we analyzed GSA’s leasing process, reviewed lease documentation, and interviewed key GSA staff for selected leases. We also conducted interviews with officials from all 11 GSA regional offices, as well as the two tenant agencies with the most leases in our sample—the Department of Justice and Social Security



Administration—regarding their experiences with the GSA leasing process.4 In addition, we reviewed GSA leasing policies and guidance and interviewed officials from GSA headquarters about the lease procurement process. Further, we analyzed documentation and estimated the actual costs over the active terms of 11 selected GSA leases—one from each of GSA’s regions. To begin, we reviewed GSA’s publicly available inventory of leased property to understand what data were easily accessible and to determine which criteria were available for consideration. After downloading the October 2014 version of this GSA data file on November 20, 2014, we reviewed related documentation and interviewed relevant GSA officials to assess the appropriateness of this data and found them to be sufficiently reliable. We then identified criteria for selection of leases for inclusion, ensuring that we had a range of leases with different characteristics from the overall lease population of 8,510 leases that would serve as a useful sample of GSA leased properties for this engagement. These initial criteria were:

• type of property identified as 100 percent office space;

• latest lease action identified as “New” or “New/Replacing”;

• current annual rent of either (a) equal or greater than the 2014 Prospectus threshold of $2.85 million (“high value” leases) or (b) greater than $500,000 and less than $2.85 million (“standard leases”);

• lease effective date of either (a) between January 1, 2000 and December 31, 2005 or (b) between January 1, 2010 and October 31, 2014;

• GSA regional office overseeing lease; and

• current annual rent divided by rentable square footage.

4These two agencies account for 6 of the 11 leases the engagement team reviewed; four of these were GSA leases on behalf of the Department of Justice and 2 were on behalf of the Social Security Administration.



After identifying the criteria for selection of leases for analysis, we used the criteria to categorize the GSA data into four sets:

• high-value leases with lease effective date between January 1, 2000 and December 31, 2005 and current annual rent equal to or greater than $2.85 million;

• standard leases with lease effective date between January 1, 2010 and October 31, 2014 and current annual rent greater than $500,000 and less than $2.85 million;

• high-value leases with lease effective date between January 1, 2000 and December 31, 2005 and current annual rent equal to or greater than $2.85 million; and

• standard leases with lease effective date between January 1, 2010 and October 31, 2014 and current annual rent greater than $500,000 and less than $2.85 million.

For each GSA region in each of these four data sets, we then identified the GSA lease with the highest cost per square foot for potential further cost analysis.5 Using information requested and received from GSA on the 40 remaining leases, we identified the tenant agencies for which GSA has leased the properties. To finalize the list of 11 GSA leases—one from each GSA region—to be analyzed, we took steps to apply additional criteria to be applied to the list of 40 leases. This step was taken to ensure (1) that the final list included at least 5 high-value leases and at least 5 standard leases, (2) that the leases ultimately selected were not concentrated in only a few federal agencies or departments, and (3) that the final list was broadly representative of GSA leases in general. In addition, we considered the inclusion of leases with and without termination rights to ensure that leases with the active possibility of the invocation of these rights were included. The final list of 11 GSA leases analyzed—which are spread among 6 federal agencies— is shown in table 1. After identifying the 11 leases to be analyzed, we requested relevant documentation on each of them from GSA, which we reviewed to develop specific questions for regional GSA officials knowledgeable about the history and details of each individual lease.

5Four of the 11 GSA regions did not have high-value leases which met all of our selection criteria across our 4 data sets, thus there were only 40 GSA leases selected for potential inclusion in our analysis instead of 44.



Table 1: Eleven GSA Leases Selected for Analysis of GSA Leasing Cost Estimates

Region City Address State Square Feet Annual Rent 1 Cambridge 10 Fawcett Street MA 14,664 $670,181 2 Newark 11 Centre Street NJ 247,067 $13,408,867 3 Pittsburgh 3311 East Carson Street PA 95,634 $3,537,510 4 North Charleston 3950 Faber Place SC 65,124 $2,984,466 5 Chicago 3252-3264 W. Fullerton IL 13,767 $681,560 6 Omaha 1717 Avenue H East NE 73,459 $2,143,439 7 Houston 1000 Louisiana Street TX 132,539 $4,785,387 8 Salt Lake City 5425 W. Amelia Earhart Drive UT 163,040 $6,313,249 9 Fresno 1325 Broadway CA 180,481 $6,684,404 10 Seattle 319 2nd Avenue South WA 22,695 $745,051 11 Washington 1717 Pennsylvania Avenue NW DC 16,379 $1,061,977

Source: GAO Analysis of GSA information. | GAO-16-188

In addition to developing questions for regional GSA officials on each of the 11 leases after receiving lease documentation from GSA, we determined which lease cost milestones were appropriate points of comparison to analyze estimates over the GSA leasing process. To determine how GSA’s lease cost estimates change over time, we interviewed GSA officials from all 11 regions in February and March 2015 to determine when in the lease procurement process cost estimates are developed and when cost estimates are provided to the tenant agency for review. In these discussions, GSA regional representatives identified three milestone points in the leasing process when written estimates of a particular lease’s costs are provided to the tenant agency. These milestones are (1) the end of requirements development phase, when tenant agencies authorize GSA to move forward with lease procurement; (2) at lease signing, when the tenant agency agrees to the terms of the lease; and (3) when the tenant agency takes occupancy of the leased space and when GSA accepts the space and rent payments begin. GSA regional representatives told us that an occupancy agreement detailing estimated lease costs and lease terms is shared with a tenant agency at each of these points, and we contacted officials from each of GSA’s 11 regions to request copies of all of the occupancy agreements associated with each of the 11 leases being analyzed. The GAO team reviewed these agreements to determine which were associated with each of the three milestone points and, when the documentation did not include all three of the milestones identified as key points of the process, we also utilized information in supplementary lease documentation to identify cost



estimates made at the three milestone points and associate each with the related agreement. For example, regional GSA officials were unable to provide copies of agreements from the initial requirements development phase for 5 of the 11 leases. For these cases, we identified the estimates included in either a lease’s acquisition plan or the prospectus documentation for high value leases as alternative cost estimates or milestones. For leases with multiple occupancy agreements associated with a single GSA lease, we considered the circumstances based on conversations with regional officials. In one case, we examined the costs estimates based on weighted average costs, on a rent per square foot basis, for the entire lease. In another case, we ensured that we did not include cost estimates that included additional space that was not considered in the crafting of initial cost estimates because GSA could not reasonably have anticipated the tenant agency’s desire for additional space or the availability of adjacent space in the market.

To analyze federal agencies’ leasing costs over time, we compared the terms and costs for the 11 selected leases across the three milestone points: the initial cost estimate, the cost estimate at the time of lease execution, and the final cost when the agency takes occupancy of the space. GSA’s occupancy agreements with tenant agencies identify a “charge basis” on each agreement, which is the total rentable square feet associated with the agreement and, because this number is the square footage metric that GSA presents to its tenant agencies, we used this as well. In order to compare leases with different terms, the rent per rentable square foot for the first stabilized year of occupancy for each lease was calculated. This calculation allows a lease estimate that began in January to be compared to a lease estimate that began in June; both were compared on a 12-month, annualized, basis. The line items that are included in our review of cost estimates included shell rent, amortized tenant improvement costs—both general and custom—and operating costs.6 Real estate tax costs were also considered, if applicable. The total

6GSA guidance states that shell rent is the single most important component of the lease contract rent and that lessors are to refer to the GSA definition when developing their shell rent rate. This guidance also states that tenant agency-driven upgrades to building shell are to be separately priced from the building shell and are considered tenant improvements. Further, it states that although some of the specifics may vary from the practice of local real estate markets, GSA must require prospective lessors to price shell as nationally defined so that the distinction between shell and tenant improvements is consistently applied for all PBS tenant agencies across the country. The shell rental rate is identified within the lease in accordance with the building shell definition. The definition of building shell is included in all leases.



amount of tenant improvement costs paid by tenant agencies was calculated using the data on these costs and the interest rate information provided by GSA regional officials. For the purposes of this analysis, GAO has considered the lease costs that are paid by GSA to the private sector landlord monthly during the lease. Although lease payments are due monthly, the stated rent per square foot is the total annual cost of the monthly payments. The tenant in the lease pays these costs as a pass-through, as well as a management fee to GSA. However, the management fee is not considered in our analysis of GSA’s cost estimates because it is determined by GSA policy.

We conducted this performance audit from August 2014 to January 2016 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives.

Appendix II: Comments from the General Services Administration



Appendix III: Comments from the Social Security Administration


Appendix III: Comments from the Social Security Administration

Appendix IV: GAO Contact & Staff Acknowledgments


Dave Wise, (202) 512-2834 or [email protected]

In addition to the contact named above, Keith Cunningham, Assistant Director; Chad Williams, Analyst-in-Charge; Alex Lawrence; Mary Pitts; Amy Rosewarne; Crystal Wesco; Terence Lam; Josh Ormond; and Elizabeth Wood made key contributions to this report.

Appendix IV: GAO Contact & Staff Acknowledgments

Contact

Acknowledgements

(542237)

GAO-16-188 GSA Lease Estim

The Government Accountability Office, the audit, evaluation, and investigative arm of Congress, exists to support Congress in meeting its constitutional responsibilities and to help improve the performance and accountability of the federal government for the American people. GAO examines the use of public funds; evaluates federal programs and policies; and provides analyses, recommendations, and other assistance to help Congress make informed oversight, policy, and funding decisions. GAO’s commitment to good government is reflected in its core values of accountability, integrity, and reliability.

The fastest and easiest way to obtain copies of GAO documents at no cost is through GAO’s website (http://www.gao.gov). Each weekday afternoon, GAO posts on its website newly released reports, testimony, and correspondence. To have GAO e-mail you a list of newly posted products, go to http://www.gao.gov and select “E-mail Updates.”

The price of each GAO publication reflects GAO’s actual cost of production and distribution and depends on the number of pages in the publication and whether the publication is printed in color or black and white. Pricing and ordering information is posted on GAO’s website, http://www.gao.gov/ordering.htm.

Place orders by calling (202) 512-6000, toll free (866) 801-7077, or TDD (202) 512-2537.

Orders may be paid for using American Express, Discover Card, MasterCard, Visa, check, or money order. Call for additional information.

Connect with GAO on Facebook, Flickr, Twitter, and YouTube. Subscribe to our RSS Feeds or E-mail Updates. Listen to our Podcasts and read The Watchblog. Visit GAO on the web at www.gao.gov.

Contact:

Website: http://www.gao.gov/fraudnet/fraudnet.htm E-mail: [email protected] Automated answering system: (800) 424-5454 or (202) 512-7470

Katherine Siggerud, Managing Director, [email protected], (202) 512-4400, U.S. Government Accountability Office, 441 G Street NW, Room 7125, Washington, DC 20548

Chuck Young, Managing Director, [email protected], (202) 512-4800 U.S. Government Accountability Office, 441 G Street NW, Room 7149 Washington, DC 20548

GAO’s Mission

Obtaining Copies of GAO Reports and Testimony

Order by Phone

Connect with GAO

To Report Fraud, Waste, and Abuse in Federal Programs

Congressional Relations

Public Affairs

Please Print on Recycled Paper.

http://www.gao.gov/

http://www.gao.gov/

http://www.gao.gov/ordering.htm

http://facebook.com/usgao

http://flickr.com/usgao

http://twitter.com/usgao

http://youtube.com/usgao

http://www.gao.gov/feeds.html

http://www.gao.gov/subscribe/index.php

http://www.gao.gov/podcast/watchdog.html

http://blog.gao.gov/

http://www.gao.gov/

http://www.gao.gov/fraudnet/fraudnet.htm




gao-16-188, federal real property: gsa could decrease ... · for all nine of the leases gao studied...

Documents